The goals of this work are to establish the value of perceptual learning as a treatment strategy for amblyopia, and to evaluate a new hypothesis for the neural mechanism underlying amblyopia. We study a nonhuman primate model for the human visual system so that we can study developmental processes and treatment strategies without compromising the visual welfare of any child, and so that we can directly assess the underlying neural correlates of developmental visual disability. The proposed work addresses two general areas of investigation: the value of perceptual learning as a treatment option for amblyopia, and the idea that amblyopia is a deficit in decoding at a high level in the perceptual processing hierarchy. This work builds on our prior behavioral and neurophysiological studies showing that the encoding of visual information by the amblyopic visual system is relatively normal and cannot account for the high-order perceptual losses evident in amblyopes. Perceptual learning has recently been demonstrated to be beneficial for improving visual performance in adult amblyopes. We will establish whether or not perceptual learning is a viable, effective, long-lasting treatment option for amblyopia by providing extensive visual experience at different ages with challenging, integrative perceptual tasks, and assessing the degree of transfer of the training effects to other perceptual abilities and the relative permanence of the effects. We will also study the development of crowding and assess the effects of visual experience on the degree or critical span of crowding. Our working hypothesis is that the fundamental disorder in amblyopia is one of decoding of the encoded sensory information. We further hypothesize that the decoding deficit occurs at a high level in the visual system, probably beyond MT and V4, and similarly affects many perceptual processes. We propose to evaluate the nature of the decoding abnormality in amblyopes by measuring the perceptual template, evaluating the effects of perceptual learning on the template, and recording the effect of perceptual learning on the neurons that contribute to the perceptual decision. The proposed work employs a combination of psychophysical and physiological methods to evaluate the neural correlates of amblyopia and associated perceptual abnormalities. We study normal development as well as amblyopia to learn about the important constraints on adult perception. Our work will inform efforts to prevent amblyopia development in children and, through the development of child-friendly training tools, more effectively treat the disorder.